Patients with brain tumors are in desperate need of new therapies, such as those that inhibit new vessel formation (angiogenesis). Recent clinical testing of angiogenesis inhibitors has accentuated the need for non-invasive measures of tumor vasculature. Therefore, the long-term goal of this first-time competitive renewal application is to develop MRI contrast agent methods that efficiently evaluate the clinical potential of anti-angiogenic therapies. The general hypothesis is that MR contrast-agent methods will provide relevant markers of tumor angiogenesis if the biophysical relationships underlying these methods are well characterized. Excellent progress has been made in this regard, especially considering that the initial 5 year proposal was awarded 3 years of funding. Simulations and experiments, made in a rat brain tumor model, using contrast-agent T1 methods demonstrate that an accurate measurement of tissue blood volume fraction depends profoundly on the choice of imaging sequence and parameters. Studies to characterize the susceptibility-based blood volume measurements have shown that the susceptibility calibration factor is different for tumor and normal brain tissue, a new finding that may be due to the differences in vascular geometry. Treatment of the rat 9L gliosarcoma with the steroid dexamethasone demonstrated a vessel-size selective effect, which may parallel the balance of angiogenic factors. Using a novel GE(gradient-echo)/SE(spin-echo) imaging method we demonstrated that dynamic susceptibility contrast (DSC) measures of total and microvascular rCBV (relative cerebral blood volume), along with vessel size information could be obtained from patients with brain tumors. Total rCBV and vessel diameter information ware found to be statistically different between low and high-grade tumors. Discerning this difference depends on the proper consideration of contrast agent extravasation effects. The specific aims are logical and exciting extensions of the initial aims. We will continue to develop and validate the susceptibility contrast methods for measuring blood volume and vessel diameter with i. the development of a novel tumor-specific simulation model and ii. MRI and histology measurements made in a rat 9L gliosarcoma model (Aim 1). The usefulness of these methods to track changes with therapy will be evaluated (Aim 2). Tumor-appropriate methods to measure cerebral blood flow (CBF) will be developed and validated (Aim 3). The optimized CBV, CBF methods will be applied to brain tumor patients and correlated with relevant immunohistochemical markers (Aim 4). Significance: Completion of these studies should move us closer to the ultimate goal of dramatically improving the diagnosis and management of patients with vascular tumors such as gliomas. Novelty: A key unique component of the proposed studies is the characterization of the biophysical relationships underlying the proposed methods. This will not only aid in assessing the accuracy of the techniques, but will also help us to exploit the wealth of information that can be derived from such measurements.